Downloads · 30 days
3
10% of all-time downloads
parakeet-tdt-0.6b-v3-coreml-ane
beshkenadze/parakeet-tdt-0.6b-v3-coreml-ane
parakeet-tdt-0.6b-v3-coreml-ane is a machine learning model from beshkenadze. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for coreml. The card lists the license as cc-by-4.0.
The Conformer encoder of nvidia/parakeet-tdt-0.6b-v3, converted to CoreML so it runs on the Apple Neural Engine (ANE). Pair it with the MLX TDT decoder in mlx-audio-swift: the encoder runs on the ANE while decoding st…
All-time downloads
30
Public
Repo size
1.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin1.2 GB · 100%
At a glance
- Library
- coreml
- License
- cc-by-4.0
- Access
- Public
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026
- SHA
- 1f131359
Base models
Model card
From the Hugging Face model README
Parakeet-TDT-0.6b-v3 Conformer encoder — CoreML / Apple Neural Engine
The Conformer encoder of nvidia/parakeet-tdt-0.6b-v3, converted to CoreML so it runs
on the Apple Neural Engine (ANE). Pair it with the MLX TDT decoder in
mlx-audio-swift: the encoder runs on the ANE
while decoding stays on the GPU/CPU.
This is the encoder only — you still need the MLX model
(beshkenadze/parakeet-tdt-0.6b-v3-mlx-fp16) for the decoder.
Format
- CoreML MLProgram, fp16 weights, fp32 I/O,
CPU_AND_NE. - Fixed input shape:
features [1, 128, 1000](1000 mel frames ≈ 10 s) →encoded [1, 1024, 125]. A fixed shape is required for ANE residency (a dynamic axis drops it to 0%); chunks are padded to 1000 frames and the output cropped back.
Usage (mlx-audio-swift)
mlx-audio-swift-stt \
--model beshkenadze/parakeet-tdt-0.6b-v3-mlx-fp16 \
--audio input.wav --output-path out \
--coreml-encoder parakeet_enc_0.6b_v3.mlpackage \
--chunk-duration 9.95
Keep --chunk-duration ≤ 10s (the fixed encoder length).
Measured (M1 Max, TED-LIUM 3 talk, 20.8 min)
| all-MLX | hybrid (this encoder) | |
|---|---|---|
| ANE residency | — | 100% (0 graph interruptions) |
| WER vs reference | 7.28% | 7.11% (agreement 1.07%) |
| RTF (Swift release) | ~95× | ~131× (~1.38×) |
| GPU power | 17.3 W | 3.0 W (÷5.8) |
The transcript is reproduced ~1:1; CoreML-fp16 is actually closer to fp32 than the
shipped MLX-bf16 encoder. Uses only public MLModel + MLComputeUnits APIs.
Conversion
Produced with the converter in
mlx-audio-swift tools/coreml-ane/:
NeMo encoder → torch.jit.trace (fixed shape) → coremltools (fp16 MLProgram,
CPU_AND_NE).
License follows the base model (nvidia/parakeet-tdt-0.6b-v3, CC-BY-4.0).
- Library
- coreml
- License
- cc-by-4.0
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026